creating a negative class in resnet50 classification model

i am working on resnet50 model, and i would like to create a negative class for images that do not fit in the trained classes of the model. any ideas?

 采纳的回答

yes,sir,may be use the score,if score is smaller,such as < 0.1,then make the image as Nan class,and save them to analysis.

更多回答(0 个)

类别

帮助中心File Exchange 中查找有关 Deep Learning Toolbox 的更多信息

产品

版本

R2022a

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by